minimax-m2
minimax-m2
MiniMax M2 is a 230B MoE (10B active) model built for coding and agentic workflows
Memory Requirements
To run the smallest minimax-m2, you need at least 121 GB of RAM.
Capabilities
minimax-m2 models support tool use and reasoning. They are available in gguf and mlx.
About minimax-m2
Meet MiniMax-M2, a Mini model built for Max coding & agentic workflows.
MiniMax-M2 is a Mixture of Experts (MoE) model (230 billion total parameters with 10 billion active parameters) built to excel in coding and agentic tasks, all while maintaining powerful general intelligence. With just 10 billion activated parameters, MiniMax-M2 provides the sophisticated, end-to-end tool use performance expected from today's leading models, but in a streamlined form factor that makes deployment and scaling easier than ever.
Use in LM Studio
Run MiniMax-M2 locally in LM Studio on Mac with MLX, or PC with GGUF (llama.cpp). MiniMax's tool calling format is natively supported as of LM Studio 0.3.31.
Leverage the model's tool calling and agentic abilities with MCP servers via the chat UI, or tool calling via our TS or Python SDKs.
From MiniMax's model card on Hugging Face:
Highlights
Superior Intelligence. According to benchmarks from Artificial Analysis, MiniMax-M2 demonstrates highly competitive general intelligence across mathematics, science, instruction following, coding, and agentic tool use. Its composite score ranks #1 among open-source models globally.
Advanced Coding. Engineered for end-to-end developer workflows, MiniMax-M2 excels at multi-file edits, coding-run-fix loops, and test-validated repairs. Strong performance on Terminal-Bench and (Multi-)SWE-Bench–style tasks demonstrates practical effectiveness in terminals, IDEs, and CI across languages.
Agent Performance. MiniMax-M2 plans and executes complex, long-horizon toolchains across shell, browser, retrieval, and code runners. In BrowseComp-style evaluations, it consistently locates hard-to-surface sources, maintains evidence traceable, and gracefully recovers from flaky steps.
Efficient Design. With 10 billion activated parameters (230 billion in total), MiniMax-M2 delivers lower latency, lower cost, and higher throughput for interactive agents and batched sampling—perfectly aligned with the shift toward highly deployable models that still shine on coding and agentic tasks.
License
MiniMax-M2 is provided under the MIT license.